Service Level Agreement
This Service Level Agreement ("SLA") describes the support, availability, and service-credit commitments MrScraper (operated by ScrapeGPT Inc.) provides for its paid, generally available Service. It forms part of, and is governed by, the MrScraper Terms of Use.
1. Definitions
- "Service": MrScraper's paid, generally available scraping and extraction platform, including its API and dashboard.
- "Incident": any unplanned event that degrades or interrupts the Service, whether detected by our monitoring or reported to support.
- "Downtime": minutes during which the Service is unavailable, excluding the exclusions listed in Section 4.
- "Uptime": the percentage of minutes in a calendar month during which the Service is available, calculated as (Total Minutes − Downtime) ÷ Total Minutes × 100.
2. Support
We provide technical support to customers through:
- Email support
- A dedicated support channel
Standard support is available Monday to Friday, 9:00 AM to 6:00 PM UTC, excluding public holidays. For Critical (P0) incidents, priority support is available 24/7, regardless of business hours. Higher-tier plans may include a named support contact as described in the applicable order.
3. Incident Response
An "incident" is any unplanned event that degrades or interrupts the Service. Incidents are detected by our automated monitoring and alerting across the scraping and extraction API, the asynchronous job-processing pipeline, and the underlying database, cache and authentication layers, as well as through reports you submit to support. Each incident is assigned a severity based on its impact on the Service, and we respond within the target times below. "Response" means our acknowledgement and the start of investigation, not resolution.
Success rate is the percentage of scrape requests that complete successfully over a rolling window. It is the primary signal we use to classify availability incidents: the lower the success rate, the higher the severity, as shown below.
| Error Severity Level | Description | Service | Response Time | Assignment |
|---|---|---|---|---|
| P0 Critical | A full outage or risk of data loss affecting most customers, with no workaround, on the scraping and extraction API, job-processing pipeline, primary database, or authentication. | 0% to 10% success rate | 15 minutes, 24/7 | On-call engineer immediately, escalated to the engineering lead. |
| P1 Major | Major functionality is significantly degraded, but the Service remains partly usable or a workaround exists, such as API or dashboard latency and errors, delayed scheduled jobs and alerts, or cache-layer degradation. | 10% to 50% success rate | 4 hours, 24/7 | On-call engineer, escalated to the owning team. |
| P2 Minor | Limited-impact, cosmetic, or single non-critical issues that do not materially affect core scraping, such as dashboard and UI defects, request or response formatting issues, documentation, or a non-critical endpoint. | 50% to 80% success rate | 2 business days | Support team, scheduled with the owning team. |
A success rate above 80% is considered normal operation and does not constitute an incident, though we continue to monitor and improve performance.
Incident lifecycle
We manage every qualifying incident through the following stages:
- Detect and alert: automated monitoring pages our on-call engineer when availability, error-rate, or latency thresholds are breached; customer reports are triaged on receipt.
- Triage and classify: we confirm the impact and assign a P0/P1/P2 severity, escalating to additional responders for P0 and P1.
- Communicate: for P0 and P1 incidents we post, and notify affected customers. Updates are provided at least every 30 minutes for P0 and every 2 hours for P1 until the incident is mitigated.
- Mitigate and resolve: we restore service, then continue root-cause investigation until fully resolved.
- Post-incident review: for every P0 (and significant P1) incident we complete an internal post-incident review to identify root cause and preventive actions.
4. Availability
Availability Commitment
We target Service availability of 99.9% per calendar month for the generally available production Service.
Exclusions
The Uptime commitment does not apply to unavailability caused by:
- Scheduled or emergency maintenance for which we provide reasonable notice;
- Factors outside our reasonable control, including force majeure events and internet or network failures beyond our systems;
- Your equipment, software, network connections, or misconfiguration, or use of the Service in breach of the Terms of Use;
- Third-party services, upstream providers, or target websites not operated by us;
- Beta, trial, preview, or free-tier features, which are provided "as is."
Response to Downtime
If the Service is down, we will provide a root cause analysis and keep you updated at reasonable intervals until the issue is resolved.
While an incident is ongoing, we communicate its status to affected customers at reasonable intervals through our support channels, so you know the current impact and the expected path to resolution. Our priority throughout is to restore the Service as quickly as possible.
Once the Service is restored, we continue to investigate until we have identified the underlying cause. For significant incidents, we provide a root cause analysis describing what happened, the scope and duration of the impact, the actions taken to resolve it, and the measures we will take to reduce the likelihood of recurrence. If you have questions about a specific incident, you can contact us at support@mrscraper.com.